1. Filed of the Invention
This invention generally relates to a biosensor for measuring the concentration of organic molecules in a solution, and more particularly to an implantable glucose monitoring device using a glucose sensitive hydrogel and a pressure transducer, the device being proportionally responsive to increases in glucose levels in the blood when it is implanted.
2. Description of Related Art
Diabetes is one of the major diseases in the United States. In 1995, there were approximately sixteen million Americans suffering from diabetes, including those undiagnosed. It is estimated that 650,000 new cases are diagnosed each year. Diabetes was the seventh leading cause of the death listed on U.S. death certificates in 1993, according to the National Center for Health Statistics. There are two major types of diabetes: type I diabetes (10% of diabetes cases in the United States), and type II diabetes (90 % of diabetes cases in the United States). Type I diabetes is caused by an insulin deficiency due to the destruction of the pancreatic beta cells, and requires daily treatment with insulin to sustain life. Type II diabetes is caused by target organ insulin resistance resulting in a decreased responsiveness to both endogenous and exogenous insulin, and is usually managed by diet and exercise but may require treatment with insulin or other medication. Most people diagnosed with type II diabetes are over 40 years old.
Diabetes disturbs the body's ability to control tightly the level of blood glucose which is the most important and primary fuel of the body. Insulin is a critical hormone needed to keep glucose concentrations within very narrow physiological limits in normal people though high levels of carbohydrates may be consumed. Not only is insulin secreted by the beta cells of the pancreas, but also its levels are rapidly regulated by glucose concentrations in the blood. Insulin allows the passage of glucose into the targets cells which contain receptors for uptake of glucose. Diabetic patients with an elevated glucose level in the blood, hyperglycemia, have either an insulin deficiency or a decreased responsiveness to insulin.
Hyperglycemia adversely affects other physiological processes. For example, hyperglycemia causes severe water loss and dehydration. Water loss can be so severe that it decreases blood pressure, and the reduced blood pressure may lead to brain damage. As discussed in National Diabetes Data Group, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, "Diabetes In America," 2nd edition, NIH Publication No. pp. 95-1468, (1995), patients of diabetes are often subject to destructive alterations of other physiological processes, causing blindness, heart attack, stroke, periodontal disease, neuropathy, nephropathy, and atherosclerosis resulting from hyperglycemia. Tissue damage can be so extensive that amputations are required to save the patient. Also, there is always the danger in diabetics of hypoglycemia due to diet, with an insulin injection required to bring the blood glucose level back up to normal. Hypoglycemic episodes can occur without the diabetic patient being aware of it. It is required to maintain a balance between insulin injection and glucose consumption to prevent hypoglycemia. However, the condition is not fatal if proper care is taken.
In treating diabetic patients, the aim is to tightly regulate the plasma glucose level within the normal physiological range (80-120 mg/dL), so that diabetic adverse effects can be avoided. Self-monitoring of blood glucose levels using dry chemical strips with a single drop of blood is considered a major advance in diabetes management. This in vitro method of monitoring of blood glucose has two main disadvantages. The first is that sampling of blood is associated with the risk of infection, nerve and tissue damage, and discomfort to patients. The second disadvantage is the practical limitation in self-monitoring which arises because the sampling frequency is not great enough for tight control of blood glucose levels close to normal ranges over a 24-hr period. Thus, as an aid to diabetes therapy, continuous monitoring of blood glucose concentrations in vivo has long been recognized as a major objective as a future tool in the fight against diabetes.
During the past decade, intense effort has been directed toward the development of glucose monitoring biosensors as an aid to diabetes therapy. Development of an implantable glucose sensor that is specific to glucose and sensitive enough to precisely measure glucose levels in vivo would be a significant advance in the treatment of diabetes. Such ability to more closely control blood glucose levels would help prevent of complications commonly brought on by diabetes. Such a sensor would also greatly facilitate glucose level data collection, glycemia research, and development of an insulin delivery system responsive to glucose levels in diabetic patients.
Several new implantable techniques have been developed for glucose analysis in clinical practice based on electrochemical principles and employing enzymes such as glucose oxidase (GOD) for glucose recognition. Potentially implantable glucose biosensors based on electrochemical transducers are the most highly developed, and this class of sensors can be further subdivided into potentiometric sensors, conductometric sensors, and amperometric sensors. The local pH change due to production of gluconic acid in the above reaction can be measured with a pH-selective electrode or an ion selective field effect transistor (ISFET), which is the basis of the potentiometric method. Similarly, in the conductometric method, changes in the electrical resistance due to the progress of the above reaction are measured. At present, neither the potentiometric method nor the conductometric method appears to be suitable for in vivo glucose monitoring due to: (a) interference by species other than glucose in the physiological environment; (b) low sensitivity and logarithmic dependence of the signal on the glucose concentration. A linear dependence of the signal on glucose concentration is highly desirable because of the need for repeated recalibrations over time for implanted glucose sensors. However, non-linear calibration curves can be handled reasonably well using microprocessors.
The most advanced glucose sensors for in vivo monitoring are electrochemical sensors using the amperometric technique, possibly because they do offer the possibility for a linear calibration curve. In the amperometric method, an electrode is used which produces a current proportional to the diffusional flux of hydrogen peroxide (H.sub.2 O.sub.2) to the electrode surface, or, alternatively, proportional to the diffisional flux of oxygen (O.sub.2) to the electrode surface. The electrode is surrounded by a membrane layer containing immobilized GOD. The glucose reaction catalyzed by GOD given earlier produces hydrogen peroxide and consumes oxygen. An increase in the surrounding glucose concentration should increase the diffusional flux of glucose into the membrane and increase the reaction rate within the membrane. The increase in reaction rate in turn should increase the local hydrogen peroxide concentration and decrease the local oxygen concentration within the membrane. This should lead to an increase in the current detected by a hydrogen peroxide-based electrode sensor, or a decrease in current as detected by an oxygen-based electrode sensor. The latter approach, based on detecting the oxygen flux, also requires a second oxygen-based electrode sensor located in a hydrogel without the GOD enzyme. This second electrode is used as a reference. Amperometric sensors must overcome several hurdles before they will ever be useful for commercial in vivo monitoring. Current glucose sensor designs appear unlikely to solve these difficult problems in the near future. The first hurdle arises from electrochemical interference. The analyte (whether hydrogen peroxide or oxygen) must be the only species present which produces a current at the electrode. Hence for both oxygen-based and hydrogen peroxide-based glucose sensors, an inner membrane must be used which is permeable to the analyte but impermeable to endogenous interferents. This is a difficult goal to achieve due to the heavily "contaminated" nature of blood. Secondly, for the hydrogen peroxide-based sensor, mass transfer coefficients for diffusion of glucose and oxygen into the membrane containing GOD must not change with time due to an adsorbed layer. Thirdly, for both types of amperometric sensors, GOD must not deactivate with time. In clinical studies of the hydrogen peroxide-based sensor, a decay in sensitivity over the implant period was observed, a phenomenon which could not be explained by blockage of the sensor surface by protein. One possible explanation for the loss of sensitivity is hydrogen peroxide mediated GOD deactivation. For the oxygen-based sensor, this can be avoided by co-immobilizing catalase with GOD, because catalase consumes hydrogen peroxide. Fourthly, a shortage of oxygen relative to glucose can place an upper limit on the biosensor's ability to measure glucose levels. This problem is called the "oxygen deficit".
In addition to the biosensors described above, several glucose release mechanisms have been developed to release insulin directly into a diabetic's bloodstream in response to high glucose levels. One approach is to use a hydrogel having immobilized GOD, the hydrogel swelling with increases in glucose concentration using essentially the same physical phenomenon which will be employed in the glucose biosensor, described below. The amount of swelling in the insulin delivery devices was used to control insulin permeability through a hydrogel membrane. For example, as discussed shortly, the proposed biosensor infers changes in glucose concentration from changes in osmotic pressure. The osmotic pressure of a hydrogel in a confined space should have a much stronger dependence on the degree of swelling (and, indirectly, on the glucose concentration) than the rate of permeation through the gel.
The prior art teaches glucose biosensors that utilize the GOD enzyme in a hydrogel, but the prior art relies on direct electrode measurement of the resulting chemical reactions. The prior art does not teach the measurement of the glucose-induced swelling of the hydrogel as a method of measuring glucose concentrations. The prior art specifically does not teach the use of a transducer to measure hydrogel swelling in response to increases in glucose levels in the blood, the use of the transducer providing a measurement tool that avoids the problems encountered by the prior art, described above. The present invention fulfills these needs and provides further related advantages as described in the following summary.